Neonatal Encephalopathy

Predicting a newborn brain disorder using Machine Learning. The challenge is that data is very scarce.

Neonatal encephalopathy (NE) is a newborn brain disorder, often due to birth-related hypoxia or ischemia, that can lead to severe neurodevelopmental disability or death. Conventional fetal monitoring using cardiotocographs (CTGs) has limited predictive accuracy, highlighting the need for improved diagnostic tools.

We propose a deep learning model that distinguishes CTG patterns among infants diagnosed with NE. These models typically require extensive data for robust model training. We propose domain adaptation, which involves pretraining on larger datasets with subsequent fine-tuning on smaller, targeted datasets.